Continuous latent representations for modeling precipitation with deep learning

TL;DR

This paper introduces a deep learning approach to generate a continuous, normally distributed proxy for precipitation data, improving simulation and downscaling by addressing data discontinuities and extreme values.

Contribution

It develops a novel pseudo-precipitation variable using deep learning, enabling better representation and downscaling of precipitation data.

Findings

Successfully created a smooth, continuous precipitation proxy

Demonstrated effective downscaling from 1° to 0.25° resolution

Addressed issues of intermittency and extremes in precipitation modeling

Abstract

The sparse and spatio-temporally discontinuous nature of precipitation data presents significant challenges for simulation and statistical processing for bias correction and downscaling. These include incorrect representation of intermittency and extreme values (critical for hydrology applications), Gibbs phenomenon upon regridding, and lack of fine scales details. To address these challenges, a common approach is to transform the precipitation variable nonlinearly into one that is more malleable. In this work, we explore how deep learning can be used to generate a smooth, spatio-temporally continuous variable as a proxy for simulation of precipitation data. We develop a normally distributed field called pseudo-precipitation (PP) as an alternative for simulating precipitation. The practical applicability of this variable is investigated by applying it for downscaling precipitation from \(1\degree\) (\(\sim\) 100 km) to \(0.25\degree\) (\(\sim\) 25 km).